| | Professional edition
Academic Research edition | Education edition | Demo edition |
| Partitioning | | | |
| # Rows | Original data : 60000 Output : 60000, subject to training partition being 10000 | Original data : 60000 Output : 60000, subject to training partition being 10000 | Original data : 600 Output : 600, subject to training partition being not more than 200. |
| # Columns | Original data : No limit Output : 200 | Original data : No limit Output : 200 | Original data : No limit Output : 200 |
| Sample from worksheet | | | |
| # Rows | Original data: Max. Excel limit. Sample output: 60000 | Original data: Max. Excel limit. Sample output: 60000 | Original data: Max. Excel limit. Sample output: 200 |
| # Columns | Original data : No limit Output : 200 | Original data : No Limit Output : 200 | Original data : No limit Output : 200 |
| # categories for Stratum variable (in Stratified Sampling) | 30 (Stratum values are not case sensitive) | 30 (Stratum values are not case sensitive) | 30 (Stratum values are not case sensitive) |
| Sample from database | | | |
| # Fields | In the table: No limit Sample output: 200 | Not applicable, as the feature is not supported. | In the table: No limit Sample output: 200 |
| # Records | In the table : 10,000,000 Sample Output : 60000 | Not applicable, as the feature is not supported. | In the table : 200 Sample Output : 200 |
| # categories for Stratum field (in Stratified Sampling) | 30 (Stratum values are not case sensitive) | Not applicable, as the feature is not supported. | 30 (Stratum values are not case sensitive) |
| Handle Missing values | | | |
| # Rows | 60000 | 60000 | 200 |
| # Columns | 200 | 200 | 200 |
| #Missing values that can be treated at a time | 60000 | 60000 | 200 |
| Bin Continuous Data | | | |
| Sum of #Columns present in the data range and #columns selected for binning | 200 | 200 | 200 |
| # Rows | 60000 | 60000 | 200 |
| # Columns in the output | 200 (Inclusive of all columns in the data range and binned columns) | 200 (Inclusive of all columns in the data range and binned columns) | 200 (Inclusive of all columns in the data range and binned columns) |
| Transform Categorical Data | | | |
| #Rows | 60000 | 60000 | 200 |
| # Columns | 200 (Inclusive of all columns in the data range and the ones added in the output.) | 200 (Inclusive of all columns in the data range and the ones added in the output.) | 200 (Inclusive of all columns in the data range and the ones added in the output.) |
| #distinct classes | 30 | 30 | 30 |
| #output variables | 30 | 30 | 30 |
| Time Series | | | |
| #Rows | 10000 | 10000 | 200 |
| Classification and Prediction | | | |
| # Rows | 10000 for Training 60000 for Training + Validation + Test (if partitioning is used) 60000 in new data used as Scoring target | 10000 for Training 60000 for Training + Validation + Test (if partitioning is used) 60000 in new data used as Scoring target | 200 for each partition (Training, Validation, Test) if partitioning is used. 200 if partitioning is not used. 200 in new data used as Scoring target |
| # Columns (input variables) | 30 (The data set can contain up to 200 columns, out of which, up to 30 can be selected for the model as input variables) | 30 (The data set can contain up to 200 columns, out of which, up to 30 can be selected for the model as input variables) | 30 (The data set can contain up to 200 columns, out of which, up to 30 can be selected for the model as input variables) |
| # Distinct classes for a categorical variable | 30 (Class values are not case sensitive) | 30 (Class values are not case sensitive) | 30 (Class values are not case sensitive) |
| # Distinct values for any input variable for Naive Bayes classification | 1000 (Values are not case sensitive) | 1000 (Values are not case sensitive) | 30 (Values are not case sensitive) |
| # Nearest neighbors for k-Nearest Neighbors | 20 (or # Training rows whichever is smaller) | 20 (or # Training rows whichever is smaller) | 20 (or # Training rows whichever is smaller) |
| # Splits for Regression Tree | 5000 (or [# Training rows -1] whichever is smaller) | 5000 (or [# Training rows -1] whichever is smaller) | 5000 (or [# Training rows -1] whichever is smaller) |
| # Levels in Tree drawing for Regression and Classification trees | 7 (Actual tree may contain more levels) | 7 (Actual tree may contain more levels) | 7 (Actual tree may contain more levels) |
| # Epochs for Neural Networks | 60000 | 60000 | 200 |
| # Iterations for Logistic Regression | 100 | 100 | 100 |
| Affinity - Association Rules | | | |
| # Transactions | 60000 | 60000 | 200 |
| # Distinct items in data set | 5000 | 5000 | 1000 |
| # Items in a transaction | 30 | 30 | 30 |
| # Rules | 60000 (Additional rules may exist, they are not displayed) | 60000 (Additional rules may exist, they are not displayed) | 60000 (Additional rules may exist, they are notdisplayed) |
| Data Exploration & Reduction | | | |
| # Rows | 20000 Exception: when using Hierarchical Clustering, the number of rows is limited to 4000. | 20000 Exception: when using Hierarchical Clustering, the number of rows is limited to 4000 | 200 |
| # Columns (variables) | 200 | 200 | 30. (The data set can contain up to 200 columns, out of which, up to 30 can be selected as variables for the model .) |
| # Clusters displayed in a Dendrogram | 30 (The solution may involve a higher number of clusters, but the Dendrogram shows a maximum of 30 top-level clusters) | 30 (The solution may involve a higher number of clusters, but the Dendrogram shows a maximum of 30 top-level clusters) | 30 (The solution may involve a higher number of clusters, but the Dendrogram shows a maximum of 30 top-level clusters) |
| Size of Distance Matrix (if specified) for Hierarchical Clustering | 200 x 200 | 200 x 200 | 30 x 30 |
| # Clusters for k-Means clustering | 20 (or # Training rows whichever is smaller) | 20 (or # Training rows whichever is smaller) | 20 (or # Training rows whichever is smaller) |
| # Iterations for k-Means clustering | 50 | 50 | 50 |
| Charts | | | |
| # Rows | 10000 | 10000 | 200 |
| # Columns | Original Data : 200 For charts drawing : 5 (For Box & Matrix plots) | Original Data : 200 For charts drawing : 5 (For Box & Matrix plots) | Original Data : 200 For charts drawing : 5 (For Box & Matrix plots) |
| # Distinct values X-variable can take | 5 (for Box plot) | 5 (for Box plot) | 5 (for Box plot) |
| General | | | |
| # Worksheets in workbook (Excel File) | 245, before running any XLMiner™ procedure (Count includes any hidden/very hidden sheets also which may be present in the workbook) | 245, before running any XLMiner™ procedure (Count includes any hidden/very hidden sheets also which may be present in the workbook) | 245, before running any XLMiner™ procedure (Count includes any hidden/very hidden sheets also which may be present in the workbook) |
| Model Storage and Scoring | Included (via XLMcalc, included with Professional ed.) | via XLMcalc (purchased separately) | via XLMcalc (purchased separately) |